Florence Prigent

Quantitative single photon emission computed thallium-201 tomography for detection and localization of coronary artery disease: optimization and prospective validation of a new technique

One hundred eight-three men underwent stress-redistribution thallium-201 myocardial perfusion tomography. After evaluation of various preprocessing filters in a phantom study, the Butterworth filter with a frequency cutoff of 0.2 cycles/pixel, order 5 (which provided optimal filter power) was used in the back projection algorithm of the patient studies. All short-axis and apical portions of vertical long-axis images were quantified by dividing each myocardial slice into 60 equal sectors and displaying the maximal count per sector as a linear profile. In a pilot group consisting of 20 normal men (less than 5% likelihood of coronary artery disease) and 25 men with coronary artery disease (greater than or equal to 50% coronary stenosis by angiography), profiles representing the lowest observed value below the mean normal profiles provided the best threshold for defining normal limits. Abnormal portions of the patient profiles were plotted on a two-dimensional polar map. The polar map was divided into 102 sectors, and sectors with a probability of greater than or equal to 80% for disease of each one of the three major coronary arteries were clustered to represent specific coronary artery territories. Receiver operating characteristic curve analysis for defect size showed that the optimal threshold for defining a definite perfusion defect was 12% for the left anterior descending and left circumflex and 8% for the right coronary artery territories. These criteria were prospectively applied to an additional 92 patients with angiographic coronary artery disease, 18 patients with normal coronary arteriograms and 28 patients with less than 5% likelihood of coronary disease. Sensitivity, specificity (in patients with normal coronary arteriograms) and normalcy rate (in patients with less than 5% likelihood of coronary artery disease) for overall detection of coronary disease were 96%, 56% and 86%, respectively. Sensitivity and specificity for identification of individual diseased vessels were, respectively, 78% and 85% for the left anterior descending, 79% and 60% for the left circumflex and 81% and 71% for the right coronary artery. These results were not significantly different from those of the pilot group. An optimized quantitative method for interpretation of stress thallium-201 myocardial perfusion tomography has been developed. Prospective application of this method indicates that the technique is accurate for the overall detection of coronary artery disease and identification of disease in individual arteries.

Myocardial perfusion imaging with technetium-99m sestamibi SPECT in the evaluation of coronary artery disease

Technetium-99m (Tc-99m) sestamibi is a new myocardial perfusion imaging agent that offers significant advantages over thallium-201 (Tl-201) for myocardial perfusion imaging. The results of the current clinical trials using acquisition and processing parameters similar to those for Tl-201 and a separate (2-day) injection protocol suggest that Tc-99m sestamibi and Tl-201 single photon emission computed tomography (SPECT) provide similar information with respect to detection of myocardial perfusion defects, assessment of the pattern of defect reversibility, overall detection of coronary artery disease (CAD) and detection of disease in individual coronary arteries. Tc-99m sestamibi SPECT appears to be superior to Tc-99m sestamibi planar imaging because the former provides a higher defect contrast and is more accurate for detection of disease in individual coronary arteries. Research is currently under way addressing optimization of acquisition and processing of Tc-99m sestamibi studies and development of quantitative algorithms for detection and localization of CAD and sizing of transmural and nontransmural myocardial perfusion defects. It is expected that with the implementation of the final results of these new developments, further significant improvement in image quality will be attained, which in turn will further increase the confidence in image interpretation. Development of algorithms for analysis of end-diastolic myocardial images may allow better evaluation of small and nontransmural myocardial defects. Furthermore, gated studies may provide valuable information with respect to regional myocardial wall motion and wall thickening. With the implementation of algorithms for attenuation and scatter correction, the overall specificity of Tc-99m sestamibi SPECT should improve significantly because of a substantial decrease in the occurrence of attenuation-related image artifacts.(ABSTRACT TRUNCATED AT 250 WORDS)

Quantification of myocardial infarct size by thallium-201 single-photon emission computed tomography: experimental validation in the dog.

To evaluate the potential advantages of thallium-201 (201T1) single-photon emission computerized tomography (SPECT) to assess myocardial infarct size in the experimental animal, six normal dogs and 14 dogs with 6 to 8 hr closed-chest coronary occlusion (eight left anterior descending and six left circumflex) were studied. Ten minutes after intravenous administration of 2 mCi of 201T1, 30 projections were obtained over 180 degrees. The dogs were killed and their hearts sliced and stained by triphenyl tetrazolium chloride (TTC). Pathologic infarct size was calculated for each slice and for the entire left ventricular myocardium as percent weight. Tomograms were quantified by automatically generating maximum-count circumferential profiles, which were compared with normal limit profiles derived from the six normal dogs. Tomographic infarct size was defined as the percentage of circumferential points falling below normal for each tomogram. SPECT and TTC infarct size on 71 slices correlated highly (mean +/- SD 27.9 +/- 23.4% and 26.7 +/- 25.3%, respectively; r = .93, p less than .001, SEE = 9.4%). To determine SPECT infarct size as percent total left ventricular myocardial weight, infarct sizes from each slice were added to one another after each was multiplied by a coefficient that reflected the contribution of that slice to the total left ventricular weight. SPECT and TTC infarct size for the entire left ventricle correlated closely (mean +/- SD 20.5 +/- 7.6% and 19.3 +/- 8.3%, respectively; r = .86, p less than .001, SEE = 4.5%). It is concluded that 201T1 SPECT is a valid method for the noninvasive assessment of experimental myocardial infarct size.

Patient motion in thallium-201 myocardial SPECT imaging. An easily identified frequent source of artifactual defect

Because TI-201 SPECT requires that patients remain in an awkward position for a prolonged time, patient motion is a potentially serious source of artifactual defects on tomographic reconstructions. Thus, a simple method was developed for detection and correction of motion from SPECT images using a Co–57 point source placed on the lower anterior chest, an area remaining in the cameras field of view throughout imaging. In the absence of motion, this point source inscribes a straight line on planar summation of the 32 projections over 180°. Movement is detected by deviation from this line. The number of pixels of motion is used to shift images so that the resultant images of the point source are linear. The method of motion detection and correction was tested in 48 consecutive patients undergoing TI-201 SPECT. The corrected and uncorrected images were reconstructed and long and short axis tomographic cuts were quantitatively analyzed using circumferential profiles of maximal counts with comparison to the lower limits of normal. Motion was detected in eight of 48 patients (17%). The amount of motion was 2 pixels in three patients and 1 pixel in five patients. Quantitative defect extent was less after correction in seven of eight patients, with a mean decrease of 71% in patients with 2 pixel motion and 44% in patients with 1 pixel motion. This corresponded with a definite reduction in the size of the tomographic defect by visual analysis, and closer resemblance to quantitatively analyzed planar images performed either before or after tomography in the same patient. Patient motion is a common problem in SPECT TI-201 imaging, which may result in major tomographic artifacts. A method is described that provides a simple practical approach for the detection of patient motion. An approach to correction of motion is also described.

Noninvasive quantification of the extent of jeopardized myocardium in patients with single-vessel coronary disease by stress thallium-201 single-photon emission computerized rotational tomography

In 22 patients with single-vessel coronary artery disease and no history of infarction, stress Tl-201 rotational tomography was used to quantify the extent of jeopardized myocardium. The vertical long- and short-axis tomograms were quantified by means of maximum-count circumferential profile analysis. The scintigraphic extent of jeopardized myocardium was expressed as the percentage of profile points falling 2.5 standard deviations below a previously established mean normal profile and was correlated to a quantitatively expressed angiographic extent of jeopardized myocardium. The extent of jeopardized myocardium varied from 1% to 55% by tomography and 8% to 50% by angiography and correlated with an r = 0.79 and a 10% standard error of the estimate. Defect intensity, reflecting the mean depth by which the abnormal points fell below the normal value of greater than or equal to 10%, was 100% specific for a coronary stenosis of greater than or equal to 70%. In conclusion, this study demonstrates that: patients with single-vessel disease have highly variable extents of hypoperfused myocardium defined by Tl-201 tomography and coronary arteriography, there is a fair relationship between angiographic jeopardy score and perfusion defects by Tl-201 tomography during exercise, and Tl-201 tomography may be used to noninvasively determine the extent of hypoperfused myocardium in coronary artery disease.

Thallium-201 stress-redistribution myocardial rotational tomography: development of criteria for visual interpretation

Despite high sensitivity and specificity for overall detection of coronary artery disease (CAD), planar stress-redistribution thallium-201 (Tl-201) scintigraphy remains suboptimal in localizing disease, because of overlap of myocardial segments. Single photon emission computerized tomography (SPECT), by providing three-dimensional representation of myocardial Tl-201, offers promise for improved localization of CAD. In 50 consecutive patients (22 normal and 28 with CAD), who underwent SPECT stress-redistribution Tl-201 imaging, we systemically developed visual interpretive criteria for perfusion abnormality on SPECT. For overall detection of disease, the best criterion for abnormality was greater than or equal to 8 sectors of moderately decreased Tl-201 uptake. With this criterion, the true positive and true negative rates for overall detection of disease were 96% and 91%, respectively. The best criterion for significant defect in the anterior or posterior coronary circulation was greater than or equal to 3 sectors of moderately decreased Tl-201 uptake. With this criterion, the true positive and true negative rates for anterior circulation disease were 71% and 100%, respectively. With respect to posterior circulation disease, the true positive and true negative rates were 100% and 50%, respectively. Regarding identification of dual circulation disease, the true positive and true negative rates were 71% and 82%, respectively.

Comparison of thallium-201 SPECT and planar imaging methods for quantification of experimental myocardial infarct size

To compare single photon emission computed tomography (SPECT) and planar thallium-201 (TI-201) myocardial perfusion imaging methods for quantification of left ventricular infarct size, 12 dogs with 6 to 8 hours of closed-chest coronary occlusion and 5 normal dogs were studied. After intravenous administration of TI-201, SPECT and three-view planar images were obtained. After the animals were put to death, hearts were sliced and stained with triphenyltetrazolium chloride (TTC) for planimetric determination of left ventricular infarct size. Infarct size on each SPECT slice and planar image was defined as the percentage of circumferential count profiles falling below the limits derived from normal dogs. Infarct size as a percentage of left ventricular mass was determined from SPECT and planar images before and after correcting for differences in myocardial slice mass from apex to base. The correlation coefficients, the concordance correlation coefficients (reflecting closeness to the line of identity), and the mean absolute deviations of the four methods versus TTC staining were 0.83, 0.77, and 5.1% (SPECT, no correction); 0.85, 0.84, and 3.7% (SPECT with correction); 0.81, 0.42, and 12.9% (planar, no correction); and 0.75, 0.49, and 10.4% (planar with correction). The regression lines did not differ from the line of identity for SPECT, whereas they differed significantly for planar imaging. Thus both SPECT and planar imaging are well suited for quantification of left ventricular infarct size. SPECT, however, appears to be superior to planar imaging, since its regression line more closely approximates the line of identity.

QUANTITATIVE INTERPRETATION OF MYOCARDIAL T1???201 SINGLEPHOTON EMISSION COMPUTERIZED TOMOGRAMS: A PROBABILISTIC APPROACH TO THE ASSESSMENT OF CORONARY ARTERY DISEASE

Probabilistic criteria for abnormality would enhance application of stress-redistribution Tl-201 rotational tomography (tomo) for evaluation of coronary artery disease (CAD). Thus, 75 pts were studied, of whom 35 had angiographic CAD (≥50% coronary narrowing) and 40 were normal (nl). 0.6 cm thick oblique and sagittal tomos were obtained on all pts. On each tomo, 60 equidistant count samples were automatically obtained resulting in 273 to 764 total myocardial sample points. Each points activity was then compared by computer to previously established corresponding mean nl values. Defect extent (E) represented % of myocardial points having a count value less than nl, and defect severity (S) was defined as the sum of the deviation of all abnl points from nl. Logistic regression analysis was done on a random sample of 38 pts (23 nls and 18 with CAD) to assess the likelihood of CAD as a function of E and S of tomographic perfusion defect. Likelihood of CAD =e,k/1+ek, where k=1.01+ 1.49S–0.208E. The validity of this model was prospectively tested in the remaining 37 pts (20 nls and 17 with CAD) by comparing the predicted and observed likelihood of CAD in four subgroups (I–IV). No significant difference (p=.1) was found between predicted and observed likelihood of CAD. I II III IV Predicted 1.9% 11% 32% 98% Observed 1.1% (1/9) 33% (3/9) 44% (4/9) 89% (9/10) Thus, a logistic model has been developed and validated that assigns a coronary disease likelihood to the quantified tomographic size of myocardial perfusion defects.

QUANTIFICATION OF MYOCARDIAL INFARCT SIZE USING SINGLE PHOTON EMISSION COMPUTERIZED TOMOGRAPHY: EXPERIMENTAL VALIDATION IN THE DOG

Although useful for diagnosis of coronary disease, the ability of single photon emission computerized tomography (SPECT) to quantify infarct size (IS) is not well defined. Ten min after 2 mCi of intravenous Tl-201, 30 30-sec projections were obtained over 180° in 6 normal (nl) dogs and 10 other dogs 8 hours after closed-chest coronary occlusion (LAD in 7 and LCX in 3). After sacrifice, hearts were sliced and stained by triphenyl tetrazolium chloride (TTC). Pathologic IS was calculated on each slice and for the entire left ventricular myocardium (LV) as % weight. 0.4 cm-thick reconstructed oblique tomograms were quantified by automatic 60-point maximum count circumferential profile analysis, and computerized comparison of each points activity to nl limit values derived from the 6 nl dogs. SPECT IS on each individual tomogram was defined as the % of circumferential points over the myocardial periphery falling below nl. SPECT and TTC IS on 62 corresponding slices with infarct correlated highly (mean±SD: 33±30 vs. 34±28, respectively, r=.87, p < .001). In order to obtain SPECT IS as % total LV, IS on each individual slice was multiplied by a different weighing factor depending on the depth of the SPECT slice along the long axis of the LV and values were subsequently added for the entire LV. SPECT and TTC IS for the entire LV also correlated highly (mean±SD: 20±8 vs. 22±7, respectively, r=0.87, p<0.01). Thus, computerized automatic analysis of SPECT T1–201 myocardial images is a valid method for the noninvasive assessment of the extent of myocardial infarction and thus may become a powerful prognostic indicator in ischemic heart disease.